Dedicated to improving quality of life by enhancing the functionality of artificial hands and their control in human-machine systems.
To this end, we study:
- Neuromuscular biomechanics of human grasp to elucidate patterns of reflex-like grip responses that can be used as inspiration for low-level reflexes in artificial hands
- Neuromuscular control of multiple digits during manipulation tasks
- Tactile sensor technology that can provide rich tactile feedback for use in real-time control of artificial fingertips
- Machine-learning algorithms that will enable the mapping of tactile sensor signals to features of finger-object interactions
- Reinforcement learning approaches to enable robots to learn how to perform hard-to-code tasks through embodied experience
- Control and sensory challenges for human-machine systems
- Reducing cognitive burden via sensory-event driven, low-level reflex algorithms. Such artificial reflexes could serve as “survival instincts” which buy time for human operators of robotic devices to detect, process, and command a response to perturbations.
Our research is intended to advance the design and control of human-machine systems as well as autonomous robotic systems.
Example applications include:
- Prosthetic hands for improving the independence and quality of life of amputees
- Wheelchair-mounted robot hand and arm for increasing the workspace, independence, and quality of life of wheelchair users
- Semi-autonomous, teleoperated manipulators for use in harsh or limited-access environments, such as for the handling of dangerous devices or high consequence materials